(070) NALIRIFOX versus FOLFOX as first-line treatment of metastatic pancreatic ductal adenocarcinoma (mPDAC): Real-world comparative overall (OS) survival analysis

Background: NAPOLI 3 (NCT04083235), a randomized, open-label, phase 3 study, has demonstrated statistically significant survival results of NALIRIFOX (liposomal Irinotecan + 5-fluorouracil/leucovorin + oxaliplatin) regimen compared with nab-paclitaxel + gemcitabine as first-line (1L) therapy in patients with mPDAC. We explored the contribution of liposomal irinotecan in the NALIRIFOX regimen by comparing OS from NALIRIFOX in NAPOLI 3 and matched FOLFOX (5-fluorouracil/leucovorin + oxaliplatin) real-world (RW) external controls.

Objectives: To compare OS between patients with mPDAC treated with 1L NALIRIFOX in the NAPOLI 3 trial and an external control arm of patients with mPDAC treated with 1L FOLFOX in RW clinical practice.

Methods: NALIRIFOX arm (N=364) was based on NAPOLI 3 safety population. RW FOLFOX arm was constructed from Flatiron Health database per NAPOLI 3 eligibility criteria. Propensity score (PS) matching with replacement in a 1:1 ratio was used to account for baseline differences between the NALIRIFOX arm and RW FOLFOX arm. PS were constructed from key prognostic factors from literature, clinical expert opinion, and stepwise variable selection, including age, gender, race, metastatic status at initial diagnosis, time from mPDAC diagnosis to 1L initiation, ECOG score, prior surgery or chemotherapy, and serum total bilirubin level within 28 days of 1L therapy. OS between the two PS matched arms was assessed using Kaplan-Meier analysis and multivariate Cox proportional hazards model for a doubly robust estimator.

Results: 33 unique patients in the RW FOLFOX arm were matched with replacement to 364 patients in the NALIRIFOX arm in a 1:1 ratio, producing an effective sample size of 364 for the RW FOLFOX arm. After PS matching, balance between the two arms was achieved for most covariates including ECOG score of 1 (NALIRIFOX: 59%, FOLFOX: 55%) and age ≥65 years (NALIRIFOX: 49%, FOLFOX: 45%). Remaining imbalances included a higher proportion of males (66% vs 53%) and a lower proportion of patients of White race (76% vs 83%) in the FOLFOX arm compared with the NALIRIFOX arm. Median OS (95% confidence interval [CI]) was 11.4 (10.2-12.2) months and 7.1 (7.1-7.9) months for NALIRIFOX and FOLFOX arms, respectively. Adjusted by remaining unbalanced baseline characteristics, NALIRIFOX was associated with a 53% lower hazard of death (hazard ratio: 0.47 [95% CI: 0.28-0.76], p=0.002) compared to FOLFOX.

Conclusions: NALIRIFOX regimen in NAPOLI 3 resulted in significantly improved OS compared to FOLFOX in a RW clinical practice setting, suggesting incremental survival benefits associated with liposomal irinotecan in the NALIRIFOX regimen. Limitations of this study include small sample size in the RW FOLFOX cohort and unmeasured confounding.